LLM Developer (Gen AI - Python) Internship

Role Overview We are seeking a highly motivated and technically sharp GenAI/Python Developer Intern to join our engineering team. In this role, you will focus on building enterprise-grade, high-performance AI systems that bridge open-source local language models and commercial APIs with core ERP workflows. You will design full-stack backend solutions that power both internal business tools and client-facing interfaces while maintaining rigorous enterprise data privacy and security boundaries.

Key Responsibilities

• Hybrid AI Pipeline Development: Design and deploy advanced GenAI architectures leveraging a mix of commercial APIs and locally hosted open-source models optimized for low latency and data containment.
• ERP Framework Integration: Develop, optimize, and maintain custom Python backend modules to seamlessly bridge complex LLM logic with Odoo's native ecosystem, database models, and relational data structures.
• Portal-Level System Deployment: Implement secure, scalable conversational layers and intelligent utility interfaces within internal employee portals and client-facing web surfaces.
• AI Safety & Alignment Engineering: Build strict input/output verification layers, semantic boundary guardrails, and automated filtering to ensure production-grade safety, data privacy, and deterministic reliability.
• Knowledge Retrieval Systems: Construct, benchmark, and maintain advanced Retrieval-Augmented Generation (RAG) pipelines over dynamic, multi-source enterprise knowledge bases.
  
  

Technical Must-Haves

• Core Language & Engineering: Mastery of Python, object-oriented programming, and clean, scalable backend architecture practices.
• ERP Ecosystem: Hands-on familiarity or a strong conceptual understanding of Odoo development (Odoo ORM, Python, XML, PostgreSQL).
• Local Inference & Deployment: Direct exposure to running local models, model quantization techniques, and inference engines such as Ollama, vLLM, or Hugging Face TGI.
• Custom LLM Orchestration: Ability to write raw Python orchestration logic for handling context windows, state management, and direct model-to-tool API execution without relying on heavy third-party agent wrappers.
• AI Safety Infrastructure: Familiarity with LLM guardrail frameworks (e.g., NeMo Guardrails, Llama Guard) or custom rule-based validation engineering techniques.
• Vector Architectures: Experience utilizing vector databases (e.g., ChromaDB, FAISS, Qdrant, or Pinecone) for high-dimensional similarity search.
  
  

Behavioral & Nice-to-Have Skills

• System-Level Vision: Ability to conceptualize full solutions—understanding how data transforms from raw user input to secure database-level mutations.
• Privacy-First Mindset: A proactive awareness of enterprise compliance, data leak prevention, and adversarial boundary testing for LLMs.
• Autonomy & Proactiveness: A sharp troubleshooting instinct with the drive to independently parse open-source papers, GitHub repos, and documentation to solve complex deployment blocks.
  
  

What We Offer

• Advanced Tech Stack: Work directly on live, privacy-centric local AI deployments instead of standard cloud-dependent APIs.
• Direct Mentorship: Collaborate closely with senior AI architects and veteran ERP systems developers.
• Career Fast-Track: Top-performing interns will be directly considered for full-time conversion opportunities as Full-Stack GenAI Engineers.